多稳态势能场驱动的无人机集群相变避障方法

何明; 陈启洋; 韩伟; 潘璠; 马一松

doi:10.11999/JEIT260357

多稳态势能场驱动的无人机集群相变避障方法

doi: 10.11999/JEIT260357 cstr: 32379.14.JEIT260357

何明^{1, 2, 3},
陈启洋^1, ,,
韩伟¹,
潘璠⁴,
马一松²

1.
陆军工程大学指挥控制工程学院南京 210007
2.
中国人民解放军32180部队北京 100072
3.
近地面探测全国重点实验室北京 100072
4.
中国人民解放军96901部队北京 100094

基金项目: 国家自然科学基金(62273356)，国家级人才项目(2022-JCJQ-ZQ-001)，国家重点研发计划(2024YFF140140)，启元实验室基金(2025-JCJQ-LA-001-101)

详细信息

作者简介:
何明：男，博士，教授

陈启洋：男，硕士生，研究方向为无人机集群指挥控制

韩伟：女，博士，研究方向为无人化指挥控制

潘璠：男，博士，副研究员，研究方向为网络信息安全

马一松：男，硕士，研究方向为指挥控制

通讯作者:
陈启洋　2300230768@qq.com

中图分类号: TP273
计量
- 文章访问数: 13
- HTML全文浏览量: 4
- PDF下载量: 0
- 被引次数: 0
出版历程
- 收稿日期: 2026-03-30
- 修回日期: 2026-06-22
- 录用日期: 2026-06-29
- 网络出版日期: 2026-07-07

A Phase Transition Obstacle Avoidance Method for UAV Swarms Driven by Multistable Potential Fields

HE Ming^{1, 2, 3},
CHEN QiYang^{1
, ,},
HAN Wei¹,
PAN Pan⁴,
MA YiSong²

1.
College of Command and Control Engineering, Army Engineering University, Nanjing 210007, Jiangsu Province
2.
National Key Laboratory of Near-Surface Detection, Beijing 100072, China
3.
Unit 32180 Chinese People's Liberation Army, Beijing 100072, China
4.
Unit 96901, Chinese People's Liberation Army, Beijing 100094, China

Funds: The National Natural Science Foundation of China (62273356), National-level Talent Program (2022-JCJQ-ZQ-001), National Key Research and Development Program (2024YFF140140), Qiyuan Laboratory Fund (2025-JCJQ-LA-001-101)

摘要

摘要: 针对动态环境下无人机集群避障过程中出现的行为模式切换不连续和控制抖振的问题，该文提出了一种多稳态势能场驱动的无人机集群相变避障方法。该方法基于局部环境风险感知构建全局风险共识机制，并通过非线性映射生成形态因子作为集群行为演化的序参量；进一步构建包含编队势、避障势与导航势的统一时变势能场模型，通过形态因子驱动势能场的连续重构，使集群不同相态对应于势能场中的多稳态势阱，设计了基于势能场负梯度的分布式一致性控制律，引入阻尼项耗散系统动能，从而将集群行为变化刻画为势能场稳态结构的连续相变过程。仿真结果表明，相较阈值切换法，集群控制输入变化率降低约26%，控制峰值降低约18%。相较仿生分流方法，平均恢复时间缩短约16 %，表明所提方法在动态环境下能够实现行为模式的连续演化与平滑控制，有效抑制抖振并提升集群整体稳定性与协同效率。
- 无人机集群 /
- 相变控制 /
- 群体智能 /
- 避障 /
- 多稳态势能场
Abstract: Objective Unmanned aerial vehicle swarms have demonstrated significant potential in complex missions such as search, surveillance, and disaster response due to their distributed coordination and robustness. However, in dynamic environments with dense obstacles and rapidly changing risks, conventional swarm control methods often suffer from discontinuous behavior switching and control chattering, which degrade stability and coordination efficiency. Existing approaches, including threshold-based switching and fixed-weight artificial potential fields, rely on abrupt transitions between behavioral modes, leading to oscillations. To address these issues, this paper proposes a phase transition obstacle avoidance method for UAV swarms driven by multi-stable potential fields, where swarm behavior evolution is modeled as a continuous phase transition process within a unified potential field framework, enabling smooth and adaptive transitions between formation and obstacle avoidance behaviors. Methods An environmental situation awareness model is first established by integrating static obstacle risk, dynamic obstacle risk, and inter-agent proximity risk. A distributed consensus protocol is employed to obtain global risk awareness. A morphology factor is then introduced as an order parameter via nonlinear mapping of the global risk, characterizing the macroscopic swarm state.A unified time-varying potential field is constructed, consisting of formation, obstacle avoidance, and navigation potentials, whose relative weights are dynamically adjusted by the morphology factor. When the risk is low, the system exhibits a mono-stable structure dominated by formation and navigation potentials; as the risk increases, it transitions to a multi-stable structure dominated by avoidance potential, enabling distributed obstacle avoidance.A distributed consensus control law based on the negative gradient of the potential field is further designed. A damping term ensures energy dissipation and stability, while a dynamic compensation term addresses nonlinear dynamics. The control law relies only on local information, ensuring scalability. The global uniform ultimate boundedness of the closed-loop system is proven using Lyapunov theory. Results and Discussions Simulation results demonstrate that the proposed method enables the swarm to maintain compact solid-phase formation in low-risk areas and smoothly transit to dispersed liquid-phase structure when encountering obstacles, followed by rapid integral regrouping after obstacle traversal. The pitch and roll angles of each UAV change gently without sharp jumps, and the distances between UAVs and obstacles/inter-UAV gaps always stay above the preset safety threshold, guaranteeing collision-free flight. Quantitative comparison statistics from 20 repeated experiments show that compared with threshold-switching control, the variation rate of control inputs is reduced by 26% and the peak control magnitude drops by 18%; compared with bionic shunting methods, the formation recovery time after obstacle avoidance is shortened by 16%. Ablation test results verify that removing the morphology-driven phase transition mechanism will obviously amplify trajectory fluctuation and control oscillation, which proves the core role of multi-stable continuous phase transition in smoothing swarm motion. In narrow channel complex environments, the proposed method effectively avoids the local minimum trap existing in traditional APF and produces smoother flight trajectories without obvious vibration. Conclusions A phase transition obstacle avoidance method for UAV swarms based on multi-stable potential fields is proposed. By introducing a morphology factor and constructing a unified potential field, swarm behavior evolution is modeled as a continuous phase transition process. The distributed control law ensures smooth behavior transitions, stability, and scalability.Simulation results verify that the proposed method outperforms conventional approaches in safety, smoothness, and coordination efficiency.
- UAV swarms /
- phase transition control /
- swarm intelligence /
- obstacle avoidance /
- multi-stable potential field

HTML全文

图 1 三维坐标系建模

下载: 全尺寸图片幻灯片

图 2 无人机避障示意图

下载: 全尺寸图片幻灯片

图 3 集群相态转换示意图

下载: 全尺寸图片幻灯片

图 4 统一势场示意图

下载: 全尺寸图片幻灯片

图 5 集群相变避障方法总体框架

下载: 全尺寸图片幻灯片

图 6 两种场景下集群避障轨迹

下载: 全尺寸图片幻灯片

图 7 多障碍物场景集群运动角度分析

下载: 全尺寸图片幻灯片

图 8 两种场景下集群避障效果分析

下载: 全尺寸图片幻灯片

图 9 轨迹质量对比

下载: 全尺寸图片幻灯片

图 10 多种方法对比分析

下载: 全尺寸图片幻灯片

1 多稳态势能场驱动的集群相变避障算法

输入：无人机$ i $状态，邻居无人机集合$ {N}_{i} $，风险权重$ \kappa $，系统参　数(安全距离$ {d}_{\mathrm{safe}} $，探测半径$ {R}_{\mathrm{S}} $，通信半径$ {R}_{\mathrm{C}} $)
输出：无人机控制输入$ {u}_{i} $
1.初始化参数：编队势，避障势，导航势函数，局部风险值，全　局共识风险值，形态因子$ {\varPhi } $;
2.while 未到达目标点：
// 局部环境风险计算
3.计算无人机与障碍物距离，与邻居距离;
4.计算静态障碍风险,动态障碍风险,机间风险;
// 全局风险共识与形态因子生成
5.执行一致性迭代，获得全局风险共识$ \overline{R}(t) $;
6.通过对全局风险公式进行非线性映射得到形态因子$ {\varPhi } $;
//统一时变势能场构建
7.通过公式(17)得到编队势$ {U}_{\mathrm{form}},{U}_{\mathrm{obs}},{U}_{\mathrm{nav}} $;
8.结合势能权重计算得到$ U(\boldsymbol{X},{\varPhi }) $;
// 负梯度一致性控制律
9.计算总势能对无人机位置$ {\boldsymbol{p}}_{i} $的负梯度方向;
10.计算无人机控制输入$ {\boldsymbol{u}}_{i} $;
// 状态更新
执行控制输入$ {\boldsymbol{u}}_{i} $更新位置$ {\boldsymbol{p}}_{i} $、速度$ {\boldsymbol{v}}_{i} $ ;
End While

下载: 导出CSV

表 1 参数设置

参数符号	物理含义	取值
$ N $	无人机数量/个	15
$ {r}_{\mathrm{v}} $	包络半径/$ \mathrm{m} $	1
$ {d}_{\mathrm{saf}e} $	安全距离/$ \mathrm{m} $	1.2
$ {R}_{\mathrm{S}} $	探测半径/$ \mathrm{m} $	8
$ {R}_{\mathrm{C}} $	通信半径/$ \mathrm{m} $	6
$ {\kappa }_{1} $	静态障碍风险权重	0.5
$ {\kappa }_{2} $	动态障碍风险权重	0.3
$ {\kappa }_{3} $	邻居风险权重	0.2

下载: 导出CSV

参考文献(22)

[1]	张欣睿, 时晨光, 吴志锋, 等. 动态威胁下基于改进APF-RRT算法的无人机集群隐身航迹规划算法[J]. 电子与信息学报, 2025, 47(12): 5178–5191. doi: 10.11999/JEIT250554. ZHANG Xinrui, SHI Chenguang, WU Zhifeng, et al. Stealthy path planning algorithm for UAV swarm based on improved APF-RRT under dynamic threat[J]. Journal of Electronics & Information Technology, 2025, 47(12): 5178–5191. doi: 10.11999/JEIT250554.
[2]	JING Xu. Research on key technologies of UAV cluster cooperative system for Internet of Things applications[J]. Journal of Control and Decision, 2024, 11(1): 26–35. doi: 10.1080/23307706.2022.2089749.
[3]	薛健, 赵琳, 向贤财, 等. 非完全信息下无人机集群对抗研究综述[J]. 电子与信息学报, 2024, 46(4): 1157–1172. doi: 10.11999/JEIT230544. XUE Jian, ZHAO Lin, XIANG Xiancai, et al. A review of the research on UAV swarm confrontation under incomplete information[J]. Journal of Electronics & Information Technology, 2024, 46(4): 1157–1172. doi: 10.11999/JEIT230544.
[4]	ZHANG Jie, LI Fengyun, LI Jiacheng, et al. A pseudo-exponential-based artificial potential field method for UAV cluster control under static and dynamical obstacles[J]. Drones, 2024, 8(9): 506. doi: 10.3390/drones8090506.
[5]	王傲, 李大鹏, 徐逸凡, 等. 先验引导的时序融合多无人机协同避障航路规划方法[J]. 电子与信息学报, 2026, 48(5): 2089–2101. doi: 10.11999/JEIT251231. WANG Ao, LI Dapeng, XU Yifan, et al. Prior-guided temporal fusion method for multi-UAV cooperative obstacle-avoidance route planning[J]. Journal of Electronics & Information Technology, 2026, 48(5): 2089–2101. doi: 10.11999/JEIT251231.
[6]	HE Yunfeng, SHI Xianjun, LU Jianhua, et al. Grouping formation and obstacle avoidance control of UAV swarm based on synchronous DMPC[J]. International Journal of Aerospace Engineering, 2024, 2024(1): 4934194. doi: 10.1155/2024/4934194.
[7]	QIU Zhenyue, ZHANG Lei, CHI Yuan, et al. Active obstacle avoidance of multi-rotor UAV swarm based on stress matrix formation method[J]. Mathematics, 2025, 13(1): 86. doi: 10.3390/math13010086.
[8]	ANWAR M A, JAVED M M, AHMAD R, et al. A digital twin approach to dynamic obstacle avoidance in UAV swarm operations[C]. 2024 5th International Conference on Emerging Trends in Electrical, Electronic and Communications Engineering (ELECOM), Balaclava, Mauritius, 2024: 1–6. doi: 10.1109/ELECOM63163.2024.10892180.
[9]	WUBBEN J, HERNÁNDEZ D, CECILIA J M, et al. Assignment and take-off approaches for large-scale autonomous UAV swarms[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(5): 4836–4847. doi: 10.1109/TITS.2023.3242765.
[10]	FAN Ruitao, WANG Jintao, HAN Weixin, et al. UAV swarm control based on hybrid bionic swarm intelligence[J]. Guidance, Navigation and Control, 2023, 3(2): 2350008. doi: 10.1142/S2737480723500085.
[11]	梅宇, 段海滨, 魏晨. 仿欧椋鸟分岔–聚集自适应控制在复杂多障碍环境中的无人机集群协同[J]. 中国科学: 信息科学, 2025, 55(12): 3114–3128. doi: 10.1360/SSI-2025-0275. MEI Yu, DUAN Haibin, and WEI Chen. Starling-inspired bifurcation-aggregation adaptive control for UAV swarms in complex multi-obstacle environments[J]. Scientia Sinica Informationis, 2025, 55(12): 3114–3128. doi: 10.1360/SSI-2025-0275.
[12]	何明, 陈浩天, 韩伟, 等. 无人机仿鸟群协同控制发展现状及关键技术[J]. 航空学报, 2024, 45(20): 29946. doi: 10.7527/S1000-6893.2024.29946. HE Ming, CHEN Haotian, HAN Wei, et al. Development status and key technologies of cooperative control of bird-inspired UAV swarms[J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(20): 29946. doi: 10.7527/S1000-6893.2024.29946.
[13]	BIDEAU P, PHAM D, DHELLEMMES F, et al. Watching swarm dynamics from above: A framework for advanced object tracking in drone videos[J]. International Journal of Computer Vision, 2026, 134(4): 146. doi: 10.1007/s11263-025-02713-y.
[14]	HUANG Jialei, ZHU Bo, and HU Tianjiang. ATI: Assemble topological interaction overcomes consistency–cohesion trade‐off in bird flocking[J]. IET Cyber-Systems and Robotics, 2024, 6(2): e12114. doi: 10.1049/csy2.12114.
[15]	ZHU Xiaoyuan, LI Yuxue, WANG He, et al. Integrated physics-data based LPV attitude control of quadrotor UAV system[J]. IEEE Transactions on Industrial Electronics, 2025, 72(9): 9635–9644. doi: 10.1109/TIE.2025.3544211.
[16]	CUI Yukang, LIANG Yiwen, LUO Qianyue, et al. Resilient consensus control of heterogeneous multi-UAV systems with leader of unknown input against Byzantine attacks[J]. IEEE Transactions on Automation Science and Engineering, 2025, 22: 5388–5399. doi: 10.1109/tase.2024.3420697.
[17]	刘航, 袁洋, 段海滨, 等. 基于变权重改进鸽群优化的无人机集群围捕控制[J]. 机器人, 2025, 47(3): 438–447. doi: 10.13973/j.cnki.robot.250003. LIU Hang, YUAN Yang, DUAN Haibin, et al. Enclosing control of UAV swarm based on the improved pigeon inspired optimization with variable weights[J]. Robot, 2025, 47(3): 438–447. doi: 10.13973/j.cnki.robot.250003.
[18]	MOUY H. Murmurations as extended strategic negotiations over roost hierarchy and safe roost positions, particularly in starlings (Sturnus vulgaris)[J]. Behavioral Ecology and Sociobiology, 2025, 79(12): 127. doi: 10.1007/s00265-025-03668-3.
[19]	POGHOSYAN S, POGHOSYAN V, ABRAHAMYAN S, et al. Cloud-based mathematical models for self-organizing swarms of UAVs: Design and analysis[J]. Drone Systems and Applications, 2024, 12: 1–12. doi: 10.1139/dsa-2023-0039.
[20]	LIAO Jian, CHENG Jun, XIN Bin, et al. UAV swarm formation reconfiguration control based on variable-stepsize MPC-APCMPIO algorithm[J]. Science China Information Sciences, 2023, 66(11): 212207. doi: 10.1007/s11432-022-3735-5.
[21]	VAIBHAV PAVLE N, RAJNEESH S, SAHOO R K, et al. Obstacle avoidance of UAV in dynamic environments using direction and velocity-adaptive artificial potential field[J]. arXiv preprint arXiv: 2512.07609. doi: 10.48550/arXiv.2512.07609. (查阅网上资料,不确定本条文献类型及格式是否正确,请确认).
[22]	WEI Jingwen. Real-time obstacle avoidance algorithms for unmanned aerial and ground vehicles[D]. [Master dissertation], The University of New South Wales, 2024. doi: 10.26190/unsworks/30514.